The use of bariatric surgery in obesity when compared with usual care does not appear to be associated with a reduction in mortality, a study of predominantly middle-aged male patients has found.

A retrospective, cohort study of bariatric surgery programmes in American medical centres from January 2000 to December 2006 was carried out.

The population studied comprised 850 patients, 74% male and 78% white, with a mean age of 49.5 years and a mean BMI of 47.4 undergoing Roux-en-Y gastric bypass.

Mortality in this group undergoing bariatric surgery was compared with that in 41,244 non-surgical controls (92% male, 68% white, mean age 54.7 years, mean BMI 42).

In patients who had bariatric surgery, the one-, two- and six-year crude mortality rates were 1.5%, 2.2% and 6.8% respectively compared with 2.2%, 4.6% and 15.2% for the non-surgical controls. Bariatric surgery was associated with a significant reduction in mortality, hazard ratio 0.64 (95% CI: 0.51-0.80). After covariate adjustment, bariatric surgery was still associated with reduced mortality but less so, HR 0.80 (95% CI: 0.63-0.995).

However, in an analysis of 1,694 propensity-matched patients, bariatric surgery was no longer associated with such a significant reduction in mortality both in unadjusted, HR 0.83 (95% CI: 0.61-1.14) and time-adjusted analysis, HR 0.94, (95% CI: 0.64-1.39).

The authors report that these findings contrast with those of previous studies which have shown survival benefits with such surgery but have studied outcomes in younger, white female populations.

They explain that obesity-related mortality is highest in men who have high rates of comorbid disease, and as such their study looked at long-term survival in such high-risk patients.

The authors also felt that their patient selection and statistical analysis was more robust than in previous studies and as such reduced the risk of significant confounding.

They acknowledge that it will be important to continue to follow the cohort to ascertain if any survival benefits from surgery appear in the longer term. It must also be remembered that this study does not include newer bariatric surgery techniques such as laparoscopic gastric banding or gastric-sleeve resections.

Bariatric surgery is an accepted intervention to reduce weight and therefore mortality in morbidly obese patients. It is currently subject to much scrutiny in our cash-deprived health service. This study suggests that it may not be as effective in reducing mortality as we believe.

In summary, this study has not shown a mortality benefit of gastric bypass bariatric surgery although this cannot be ruled out in the longer term.

It also does not consider the improvements in quality of life and comorbidities associated with the weight loss seen following such procedures.

Dr Peter Savill